Pushing Back: Living and Writing in Broken Space

Stuart Moulthrop
School of Communications Design
University of Baltimore


The word is the serpent eating its tail; it is the sign that disappears in the act of signing -- the signing is not complete until the word has disappeared into its puff of meaning. At the instant of apotheosis it ceases to be itself; when it has brokered the transaction, it vanishes, reappearing only when the eye has moved on. This is the paradox of paradoxes: The word is most signifier when it least signifies.

-- Sven Birkerts [1]

Writing should demand we see. Seeing should demand we change.

-- Fred Pfeil [2]

1: Ideas of the opposite

Critics are a blessing. As William Blake almost said, they may confer kingly titles. Failing that, critics can at least point out the high stakes involved in questioning important aspects of a culture. "Opposition is true friendship," the poet wrote, though these days that line reads less like Blake than some wistful, downsized Strangelove mourning the receding glaciers of the Cold War. There is a lesson in this. Without adversaries we are lost, unable to find even a contingent standpoint of difference in the all-assimilating postmodern funhouse. If opposition is friendship, critique is aid and comfort. It restores the agenda.

Though this logic of opposition could apply to virtually any recent intervention -- multiculturalism, or "language" poetry, or postmodern theory itself -- we focus here on an information-age practice called "cybertext" (Aarseth, 19). This is writing (or more accurately, textual production in various media) that depends on a feedback mechanism operated and partly controlled by the receiver to evoke a particular state of a variable or combinatorial text. While this formula may cover everything from libraries, encyclopedias, and aleatory poems to role-playing games, I Ching, and the Ouija board, much cybertextual interest these days concerns text production on the Internet and its World Wide Web, a subtype of electronic writing called hypertext or hypermedia.

Once upon a time, globally networked, computer-mediated communication seemed a genuinely radical notion. In the nineteen sixties and seventies, advocates like Stewart Brand, Howard Rheingold -- and above all, Ted Nelson -- foresaw electronic publishing as the cardinal technology of a de-centered, populist information culture. When this vision reached research labs at companies like SRI, Xerox, and IBM, not to mention a certain garage in Silicon Valley, it helped spur development of personal computers, which in turn nurtured the emerging Internet (see De Landa). Unlike some prophecies of the time, the development of cybertext seemed by the early eighties a viable prospect. Within ten years the visionaries were joined by designers and implementers, including Bill Atkinson, chief programmer of Apple's HyperCard, Tim Berners-Lee, inventor of the World Wide Web, and Jay David Bolter, software developer and cultural theorist. Bolter has perhaps best expressed the expected impact of cybertext on intellectual life. He predicts both an atomization and a new regime: "Although we... lose the satisfaction of belonging to a coherent cultural tradition, we gain the freedom to establish our own traditions in miniature. The computer offers people the opportunity to build liaisons with other readers and writers.... Unlike television, which promotes uniformity (even through the apparent diversity of cable and satellite stations), the microcomputer and the phone network really do permit special literacies to survive." (238).

In light of recent events, Bolter's quondam description of "network culture" seems acutely innocent, much like the utopian, "tribal" rhetoric of the sixties on the morning after Altamont. With the headlines full of pedophiles, pornographers, and saucer cultists, some consider the Internet less a pastoral retreat for readers and writers than a sinkhole of weirdness and perversion (Johnson). At the same time, the cybertextual enterprise itself has changed. When Bolter wrote Writing Space very few people were interested in commercializing the Internet. Now one cybertextual system, the World Wide Web, accounts for more than 150 million documents (a number that will seem ludicrously low by the time you read this). Profitable exploitation of this resource has become a major object of capitalist desire -- and lately, anxiety.

Bolter names our times "the late age of print" (2), but the Internet has thrust belatedness upon other media as well. As cybertext alters the conditions of writing and publishing it also unsettles two holy pillars of spectacular society, broadcasting and advertising. By inviting people to create "liaisons" at will instead of switching on a standard program, cybertext threatens to break up the broad, demographically homogeneous audiences on which traditional marketing depends. How can the logic of mass communication be preserved? One answer, according to the senior pundits of Wiredmagazine, lies in a new approach to the technology:

Sure, we'll always have Web pages. We still have postcards and telegrams, don't we? But the center of interactive media -- increasingly, the center of gravity of all media -- is moving to a post-HTML environment, a world way past a Web dominated by the page, beyond streamed audio and video, and fast into a land of push-pull, active objects, virtual space, and ambient broadcasting. You might not want to believe us, but... you can kiss your Web browser goodbye. (Kelly and Wolf, 12-13)

Though some have long suspected that HTML stands not for "Hypertext Markup Language" but rather "Hypertext More or Less," the Wired revisionists would dispense with even this distant approach to "special literacies." They would substitute the sharply different and noticeably familiar model of "push" communication. According to the "push-pull" metaphor, users of the Internet currently "pull" information into view through cybertextual interactions. This may soon change. With software now emerging, such as various "webcasting" systems, Netscape's "kiosk mode," and Microsoft's ActiveX programming, content arrives in an unbroken, often uninterruptible stream once the user completes an initial link. Since these schemes aim to make the Web safe for advertising, it is reasonable to assume that users will not be encouraged to make other connections, but rather to keep the channel open and await instructions. Don't touch that keyboard:

[The] almost neurotic urge to zap has falsely led people to think that what viewers want is more zapping, more control, more steering. What they want instead are more ways to zap. More ways of interrupting flow, more varieties of story and no-story, text and game, of things done together with other people and things done alone. More states that flit between steering the media and being steered by it. More ability to tweak the dial, between twirling and being twirled, so that finally you can dance with the media. (Kelly and Wolf, 19)

Although mention of "more ways of interrupting flow" confers some ambiguity (or inconsistency), the drift of these pronouncements seems fairly clear. If certain people have their way, tomorrow's hypertext links will be as numerous and important as today's "postcards and telegrams" -- that is, they will dwindle into obscurity. While we may never lose our Web pages, and though we may be given a thousand new ways to "zap," it seems more accurate to say that we will forever have television, albeit in new boxes. To which end we might remember what TV's Control Voice always says about trying to adjust your television set. Broadcasting, it would seem, is the re-run from which we cannot awaken, which brings us to a familiar truth: techno-cultural space curves such that no line of flight, however ambitious, ever really departs from initial conditions. Given time, any cultural revolution loops back toward reaction.

2: Brokered transactions

This is an essay about reaction -- in the large sense, reaction against changes in mass communication and mass markets, but more particularly, reaction against certain approaches to "story and no-story," or narrative. On first presentation these topics may seem dubiously related if at all. Barring a few exceptions (most notably the Miller brothers' brilliantly successful Myst), cybertextual work has little large-scale appeal or market potential. Many of the most influential and interesting examples of electronic narrative are available non-commercially on the Internet (see Shumate). So what do cybertextual stories have to do with the social and economic fate of cyberspace? More perhaps than may be apparent; but to explain the articulation it is necessary to take up some parochial controversies. As we have noted, there is much to learn from critics. Consider for instance this assessment from Jurgen Fauth:

One of the most prevalent problems plaguing current hyperfiction is what I call the "poles-in-your-face" effect: Many of the hyperstories found online are lacking in content and quality writing because the novelty of hypertext makes all other aesthetic concerns secondary. This seems to be an intrinsic problem with newly discovered forms... (par. 1)

According to Fauth, hypertext fiction writers suffer from a morbid obsession with technique. He compares hypertext links to all those toppling trees, baseball bats, monstrous limbs, telescopes, and other elongated objects of which the makers of stereoscopic movies were so fond. Links are "poles in your face" that ostensibly betray a disordered relationship between storytelling and technical effect. Following through on this analogy, Fauth like the swamis at Wired sees little future in cybertext. His Johnsonian logic runs thus: "3-D" did not last. Hyperfiction will not last. Nothing new lasts. Works like Joyce's afternoonare by implication the literary equivalent of Creature from the Black Lagoon or It Came from Outer Space -- curios, sports, and instances of bad taste, interesting only as they mark the limits of invention.

It seems wise not to touch with any sort of pole the subtext of anxiety so evident in this curious analogy; neither would it be productive to dwell on the quality of Fauth's critical judgement, which relies heavily on blanket assertion in a notable absence of close reading.[3] For present purposes it does not matter whether Fauth accurately describes the state of affairs in cybertext. Counting texts both on and off the Internet, a few hundred electronic fictions are currently available. For the sake of argument, assume that "many" if not all these productions are nearly identical and that all are so overburdened with technique that they deliver little in the way of "content and quality writing." Even under such a general indictment, questions remain.

We might begin with the critic's criterion. What could be meant in this context by narrative "content?" The term seems more reminiscent of Microsoft than the Mississippi Review. What do readers (as opposed to marketers) expect narratives to contain? Further, what aspects of a work differentiate inferior stuff from "quality writing?" In other words, to recur both to friendly opposition and poles-in-your-face, what useful ground of difference has Fauth staked out between his great tradition and the "miniature" tradition of cybertext? It is hard to answer these question based on Fauth's text, but this may be excused. After all, Fauth seems more interested in tactical (if not practical) criticism than in general theory, and his scope is the article, not the opus. His dissatisfaction with cybertext provides an opening to inquiry.

If we take up a more ambitious study, Sven Birkerts' Gutenberg Elegies, the terms of ideological division become much clearer. Though some have noted that Birkerts also stints in his engagement with electronic texts (see Kirschenbaum), he at least offers some reason for finding them unreadable. Birkerts in fact lays out quite a sophisticated theory of narrative "content," as in the first epigraph above. Sensibly enough, this theory does not explain what a narrative should contain but rather how. Content is an intangible commodity carried by the text and delivered in a "transaction" that is "brokered" by the word through its "apotheosis." The word, says Birkerts, "is the sign that disappears in the act of signing" (78). Though the comparison may not be entirely fair, this position recalls one of Donald Barthelme's characters, the numinous Miss R. of "The Indian Uprising:"

"Young people," Miss R. said," run to more unpleasant combinations as they sense the nature of our society. Some people," Miss R. said, "run to conceits or wisdom but I hold to the hard, brown, nutlike word. I might point out that there is enough aesthetic excitement here to satisfy anyone but a damned fool." (16)

To be sure, the word to which Birkerts holds seems less a dense nugget of meaning than a moveable token in a glass bead game. This is not a "self-consuming artifact" in Fish's sense of socially emergent discourse; rather, the signifying activity is self-directed, self-contained, and above all self-limiting. Words in this view (that is, printed words) signify exhaustively and absolutely. They are arranged just so, and, in the case of "quality" writing, to best effect. In this regard their status as artifact or techne has only minimal, transitory importance. Given Richard Lanham's distinction between "looking AT" and "looking THROUGH" (5) -- between uses of language that do or do not invite attention to themselves and their conditions of production -- Birkerts seems squarely on the side of transparency.

Like Fauth, Birkerts disapproves of those who would alter the business of writing by tinkering with word processors and other machines. The word obtrudes itself only so that it may vanish. Reading must be absorbing, engaging, and assiduously pursued for long periods if it is to enrich the reader through the influence of a great mind. Changes in technology can only hinder this process. Asking readers to spend hours gazing at a monitor is asking too much. Bothering the reader with hypertext links disrupts the continuous brokering of ideas. Most important, any deviation from the fixed stability of the printed page threatens the divine right of authors:

Writing on the computer promotes process over product and favors the whole over the execution of the part. As the writer grows accustomed to moving words, sentences, and paragraphs around -- to opening his lines to insertions -- his sense of linkage and necessity is affected. Less thought may be given to the ideal of inevitable expression. The expectation is no longer that there should be a single best way to say something; the writer accepts variability and is more inclined to view the work as a version. The Flaubertian tyranny of le mot juste is eclipsed, and with it, gradually, the idea of the author as a sovereign maker. (158)

Here indeed is a pole in the face, or more precisely a frontier marker announcing NO TRUE LITERATURE PERMITTED BEYOND THIS POINT. Yet even though we may have reached the border of good writing (that is, of print), its productions can still tell us something about our situation. Consider once more Barthelme's story. At the outset the formidable Miss R. appears to be some kind of psychotherapist, but by the end she is revealed as a terrorist, head of a revolutionary court that will decide the feckless narrator's fate:

"Skin," Miss R. said softly in the white, yellow room. "This is the Clemency Committee. And would you remove your belt and shoelaces." (19)

This fable comes from 1968 -- another time and a very different politics[4] -- but even so, it resonates with the present. In Barthelme's satiric view, a strict regime in language goes fist-in-glove with other deficiencies of tolerance: scratch an essentialist, find a Maoist. This gives a somewhat different spin to the "tryanny of le mot juste." When Barthelme wrote, the world seemed to be settling, as no doubt it usually is, into increasingly polarized and intransigent attitudes. Then the cry was "cultural revolution," now it is "cultural war," and though the parallel is hardly as neat as this play of words suggests, there is some ground for comparison. Sentences as harsh as Birkerts' and Fauth's leave little room for appeal, assuming one can find a Clemency Committee. Those who choose not to work in print are denied the "sovereign" status of literary artists. The "quality" writers have closed ranks around a well-worn ideology, essentially the one that Fred Pfeil described twenty years ago:

Language, like paint or musical pitches, is seen to be a self-enclosed, fanatically guarded medium. Its ability to perceive or to express perception of the world is secondary or immaterial; the desired end of the act of writing is a closed, autonomous, self-referential body of well-shaped, beautiful language. (21)

The Gutenberg elegists would no doubt object to Pfeil's claim that "perception of the world is secondary or immaterial." As they see it, the transparency of literary language allows a higher truth of human experience, refined and sanctified by the artist, to shine through. Since the nature of this higher truth (or at least its origins) is open to question, the accuracy of any such "perception" may also be dubious -- a point to which we will return. In any event the elegists would doubtless affirm a stable, closed, and autonomous literary language which, as Miss R. says, only a "damned fool" would renounce. For some damned fools, the only response to this attitude may be a hardening of opposition. Partisans of cybertext might find sympathy with John Perry Barlow's response to recent telecommunications policies:

Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace, the new home of Mind. On behalf of the future, I ask you of the past to leave us alone. You are not welcome among us. You have no sovereignty where we gather. (par. 1)

These sentiments are enormously suggestive, speaking as they do to the question of cultural "sovereignty." Could one simply secede from print culture? Is Mutual Assured Irrelevance a valid strategic doctrine for the present cultural war? Under present conditions these ideas are not as impractical as they may seem. Birkerts himself observes that conventional publishers these days are reluctant to consider books lacking mass appeal (28). Indeed, writers more interested in readership than royalties might fare better at the moment by publishing or self-publishing on the Internet than by going through traditional channels. Those concerned with professional validation could submit to an established literary or academic Web site, or more radically, could consider setting up reviewing agencies, indices, or "seals of approval" to evaluate self-published work (Ršscheisen). Various electronic payment schemes might even hold the prospect of (small) financial reward. This reasoning applies both to traditional writers as well as to those whose cybertextual work could not be reproduced in print. For the latter, of course, there are few alternatives to the Internet.[5]

But though Barlow's declaration of independence is rousingly immoderate, it seems on closer examination rather damned-foolish. Since cyberspace relies on "Industrial" ways and means, it can never really be independent. We might remind ourselves where electricity comes from. "Leave us alone" has a very nice ring, but neither cyberspace nor its textual productions are likely to be left alone by western governments, major cultural organs, or the large concentrations of capital these entities serve. The friendly opposition includes not just Fauth and Birkerts but also advocates of "push" technologies, mass marketing, and Web-as-TV. Those interested in defending alternative textual practices had better be prepared to situate those practices discursively (if not dialectically) in the non-virtual world. To return to the more parochial subject of this essay: if interactive writing matters, we must understand how, to whom, and why.

3: You can't take it to bed with you

Birkerts has many interesting things to say about the "metaphysics" of print: the relationship of writer to language, of reader to an imaginary world, and of medium and material to the process of signification. Yet when it comes to the corresponding features of cybertext, he resorts mainly to negation. Printed texts are fixed and autonomous; electronic texts are not. Traditional writers at their best are deliberate and careful; those who work with computers, he fears, may be merely prolix. Print demands perfect expression; a hypertext consists of "versions," or to borrow from Borges, "an indeterminate heap of contradictory drafts" (24). It is regrettable that Birkerts chooses not to consider cybertextual work more closely in its own right, especially since some of his claims might not hold up to testing. He asserts, for instance, that because they lack a single, determined ending, hypertexts say nothing of value about the consequences of human action. Yet many such texts are deeply concerned with causal logic and if read with reasonable engagement can convey very clear messages about choices and outcomes (see Douglas 1993).

Birkerts makes no reference to critics who have given more time to interactive texts. To be fair, some of this work is only now appearing, though a significant body of commentary on electronic writing has been available for the last five years.[6] In some ways, close attention to interactive texts confirms the suspicions of the elegiac school. Virtually all systematic accounts of such works notice their discontinuousness, convolution, and strong cognitive demands. In an aleatory poem, a "lexicon novel," or a film with multiple endings, the reader or interpreter faces an absence of familiar formal limits. With hypertext and hypermedia this problem is compounded by the convention of link-following or "navigation," which ties the discourse of the text to repeated selection or probing by the reader. Hypertext links present a particularly important concern for theorists. Some consider them barriers or "occlusions" (Glazier), others see them as gaps and fissures (Rosenberg 1995). One particularly astute reader calls them "detours," routes to destinations that never quite match our expectations (Harpold 1990, 1994).

In many hypertexts it is hard indeed to develop a sense of direction. Yet it does not follow, as the elegists fear, that writing with machines represents a wrecking of form and an assault on "sovereign" authority. Reacting with understandable dread to Robert Coover's musings on an "End of Books," Birkerts turns by association to Barthes' "Death of the Author," and levels his suspicions against technologists (152-55). But by his own admission Birkerts has spent little time reading electronic texts. Contrast the critic J.Y. Douglas, who has worked with interactive narratives, particularly hypertexts, extensively and intensively. She reaches a markedly different conclusion. Interactive construction, says Douglas, does not efface or impair the author position but rather transforms it. Authors of interactive pieces in any genre come to be what playwrights have always been: creators of initial conditions for later performance (1993, 35). Unlike the modernist revolution in form (in its theory at least), cybertext does not aim for impersonality but rather for discursive intimacy, not entrancement but engagement. Joyce's narrator says of the matrix of stories in afternoon, "there we match minds" (1990). Minds (and presumably their textual simulations) are not without structure. Cybertext is not simply a revolt against connection and coherence, but it does ask us to redefine those terms.

Consider this general premise: Cybertextual works are structures for breakdown in semantic space. This sentence contains all the difference between the elegiac position and a practical engagement with the late age of print, but since it is far from self-evident, some unpacking is in order.

The term "semantic space" comes in this instance from cognitive psychology and interface design (Dillon et al.), though it has obvious cognates in literary theory. Semantic space is an indefinite domain of possible expression to which the current text bears some active and particular relationship. Bolter calls it "a structure of possible structures" (144). As Douglas points out, there is nothing especially radical about situating writing in a virtual space: "It can be argued that since the advent of the modern novel readers have been challenged with the task of reading something that approximates the virtual, three-dimensional space of hypertext narratives" (1994, 176). Indeed an approach to text as a contingent arrangement of discursive possibilities might be traced throughout narratology from the Russian Formalist pairing of sjuzet and fabula up through structuralism, affective stylistics, and deconstruction. By the same token, non-literary rhetoric and information science have also become accustomed to thinking of texts as selections from a signifying field (Johnson-Eilola, 211-12). The fetishism of print and le mot juste among the elegists seems curiously at odds with this emphasis, though no doubt that is the point.

Semantic space is particularly useful in thinking about hypertext, with its formalized structures of discontinuity and connection. Applying the metaphor of textual "navigation," the hypertext link may be thought of as a course or channel through the space. Suppose for instance that you were reading this article on the World Wide Web instead of in print, and that the word "space" at the end of the previous sentence was configured as the cue or "yield" of a hypertext link. The click of a button would trigger a transition, replacing your view of this paragraph with, say, the cheery iconography of Heaven's Gate. But though the pivotal word literally vanishes in this instance (apotheosis or no), the effect of this transaction is far from self-effacing. Fauth's complaint about gimmickry seems borne out to some extent. The link is obtrusive and peremptory, if not a pole in your face then something equally cinematic, a quick dissolve or flash cut. Indeed, the venerable hypertext program HyperCard contained built-in visual effects to signal transitions. As nearly all commentators on hypertext observe, links dislocate even as they connect. Landow (1987) once argued for an orthopaedic rhetoric of arrivals and departures, but as Harpold and others have noticed, even the most rationally designed transition can produce something of a surprise. This is particularly true in the case of the Web, where links point to documents maintained on remote systems. Web sites follow their own erratic courses through semantic space, changing at the whim of their creators. The link that took me to the suicide cult last week may lead today to trailers for the inevitable mini-series. Here we begin to find the limits of the navigation metaphor. By contrast, the "push-pull" model may offer a more satisfactory framework. Borrowing yet again from physics (albeit the physics of Star Trek), hypertext links may be thought of as warping or wrapping semantic space around the reader's perspective (Moulthrop and Kaplan). Literally speaking, they pull distant information close. This seems closer to what goes on in the Internet, yet distant from navigation in the real world.

In actual seaborne travel, voyagers can see where they are going as they go. They pass through a tangible, substantial, and resistant medium, as their stomachs sometimes attest. Progress in Web space, on the other hand, is an experience that combines the matter transporter of science fiction with the slot machines of Las Vegas: instantaneous transition with the possibility of surprise. The time and space of passage do not directly register. With the important exception of visual games like Mystand Doom, many cybertexts (and most hypertexts) lack ways of representing a discursive vicinity. On the World Wide Web links are, in Ted Nelson's phrase, "divingboards into the darkness" (1995). Hypertext Markup Language provides very few structures for maintaining a larger context. The preterite space through or over which we move remains unseen -- though since subsequent links may take us to points within that space, it remains present to the reader's awareness at least by implication. So links may mean more than they appear to do on a simply technical level. Instead of focusing entirely on instantaneous replacement we might think of links as having two components: the visible, binary circuit of connection (technology's fort/da) and the unseen matrix or "structure of possible structures" against which this transaction is realized, figure against ground. Every link comprises both parts.

There are of course moments when ground overwhelms figure and the Web's limited implementation of hypertext acknowledges its surrounding space, albeit unintentionally. These are the familiar failures that occur when a server is too busy to respond, an address has changed, material fails to transfer quickly enough, or something else goes wrong. When these things happen, the frustrated navigator runs up against an incomplete page, an error message, or at worst a frozen screen. These untoward outcomes signify very little about the semantic space of the text, but they at least remind us we are dealing with a complex, contingent system whose behavior we cannot entirely predict. In the good old Gutenberg days, all we had to worry about were sloppy typing, printing errors, and the occasional misbound sheet. In cyberspace the possibilities for error are enormously greater, as many irritated critics note. Yet these apparent failures are more than mere annoyances. As John Milton might have said had he written from Silicon Valley, sometimes a crash can be a fortunate fall. Or to recite the credo of cybertext: That's no bug, it's a feature.

Which brings us to the second stage of the initial proposition: cybertexts are structures for breakdown. That last term must be granted a certain range of meaning. It does refer literally to moments when things go wrong in information systems, but it also indicates a more abstract and significant articulation between human and machine, or in our case, the text and its reader/performer. In fact the operation of breakdown may be the most important cultural aspect of cybertext.

The word "breakdown" comes from Terry Winograd and Fernando Flores, who borrow it from Heidegger in Understanding Computers and Cognition, a book that draws on phenomenology and the biological theory of autopoesis for a new approach to artificial intelligence and software design. Breakdown proves useful in this context because it allows Winograd and Flores to raise the intellectual stakes for programmers and engineers. Like many technologists, Winograd and Flores prefer process to product. They set aside cognitive science's dominant preference for reductive solutions, preferring to think about technologies in the phenomenological sense of articulation and unfolding. In this respect some bugs are indeed supremely important features:

...we want to break with the rationalistic tradition, proposing a different language for situations in which 'problems' arise. Following Heidegger, we prefer to talk about 'breakdowns.' By this we mean the interrupted moment of our habitual, standard, comfortable 'being-in-the-world.' Breakdowns serve an extremely important cognitive function, revealing to us the nature of our practices and equipment, making them 'present-to-hand' to us, perhaps for the first time. In this sense they function in a positive rather than a negative way. (77-78)

In the domain of cybertext, failed links and malfunctioning programs do represent cognitively significant breakdowns; but if we follow Harpold, Douglas, and other critics who insist on the deeply problematic nature of links, then we must expand this notion to cover apparently successful operations as well. If links and other interactive transactions are inherently confusing, traversing an invisible, nonspecific space, then these elements must also convey something of a phenomenological crisis or surprise even when they work as intended. Cybertextual work cannot deliver the infinite variation which its multivariate structure disingenuously promises. Indeed, no verbal structure can truly map semantic space (Dillon et al.). The reader matches wits with an articulated system or text, not another mind, and as Winograd and Flores point out, this means the outcome will always be a mismatch in the end. Texts are not minds. (For a somewhat different development of these ideas, see Moulthrop 1995.)

In this respect we can say that cybertexts are not plagued by breakdown, rather they are conceived in breakdown. Winograd and Flores point out that "[n]ew design can be created and implemented only in the space that emerges in the recurrent structure of breakdown. A design constitutes an interpretation of breakdown and a commited attempt to anticipate future breakdowns" (78). It bears noting that by "anticipate" Winograd and Flores do not mean prevent. Breakdown is an essential component of good design. Well-made software works both with and against its conceptual limits.

Drawing on these ideas, we can think of cybertexts as structures forĘbreakdown, ways of thinking critically and creatively about all the plausible, deceptive constructions we find in cyberspace. Though this logic would apply to any transaction among people, machines, and texts, the application may be clearest for hypertext links. In traversing a semantic space, the link by implication spans or contains that space, if not in its infinite totality then with a kind of cognitive blank check for which there can never be sufficient discursive funds. Links like words may be "brokers" of meaning, but they are not honest brokers. As a divingboard into darkness, the link from "space" to the saucer cult invites us to consider an enormous range of possible destinations -- from Hubble photography to differential topology to Gene Roddenberry's "final frontier." Yet only one possibility is realized, and likely as not it will not be what the reader anticipated.[7] Having followed the link, we might wonder where else we might have gone, or where other links in the same document might take us. The particular circuit just closed implies an unseen matrix of latent connections. These other possibilities are not present or realized (indeed they may be imaginary) but we cannot exclude them from the transaction. The present circuit implies an unseen machine. We perceive any given connection as figure against an undefined ground of semantic space, and in this perception we recognize that the present text is both more than we can easily perceive, but also likely far less than its complexity implies. This recognition is breakdown, or what passes for it in hypertext.

If this account does not justify cybertexts to their critics, it may at least measure the conceptual gulf between them and traditional texts. That gulf is considerable. Even when they occur in print, cybertexts cannot be well-formed or well-behaved books, and when they manifest themselves as hypertexts they pose even greater problems. Perhaps the most persistent complaint about machine-mediated texts stems from their need for bulky hardware. Books and magazines are exquisitely portable, convenient to our most intimate spaces. Likewise, television and radio have escaped from the parlor and colonized every room of the house. Meanwhile the newer media remain ponderously stranded on the desktop. "You can't take it to bed with you," the elegists still say, even in the age of laptops and digital assistants. Though the ultimate answer to this problem lies with the engineers, the concept of breakdown as a general property of cybertext suggests a more immediate response. Indeed you shouldn't take such things to bed with you -- if you go to bed chiefly for sleeping. Cybertext is not a tranquilizing agent. It is not about slumber, trance, transport, or other forms of hypnagogia. It is an irritant, often indeed an irritating enterprise, a practice full of pranks and tricks that no doubt will never last; but paradoxically these admissions do not annul its value; in fact just the opposite.

4: Better a pole in the face

"You gotta get a gimmick," Sondheim's philosophical stripper counsels, "if you want to get ahead." Though postmodernism has drummed in a deep distrust of all progress narratives, we might still find use for this sentiment even after substituting paralogy for progress. Does the gimmickry of cybertext represent a productive move in the language game of information culture? If it doesn't get us "ahead," does it get us anywhere at all? Fred Pfeil faults the school of "quality writing" for allowing "perception of the world" to become "secondary or immaterial" to its cult of fine language. With or without investment-quality writing, cybertext should also be held to this standard. Do its technical maneuvers reflect any engagement with the hard, old world of "flesh and steel?" What social effects, if any, might flow from narratives based on phenomenological breakdown? In his treatise on postmodernism, Frederic Jameson places two major items on the agenda of art and ideas: a "pedagogical political culture" and an "aesthetics of cognitive mapping" (54). It will take a movement much more ambitious and sophisticated than the present interest in cybertext to fulfill Jameson's assignment. All the same, some limited claims can be made.

The twentieth century may be marked in the final analysis less by Kuhnian "paradigm shifts" than by what Nelson with his usual keen wisdom calls "paradigm shiftlessness" (1995). That is, industrial society has been struggling for a century and more with rationalistic, theoretical, revolutionary models for everything from political economy to popular entertainment, but we have yet to escape the enduring nightmare of grand solutions, or what we might call general systems hubris. If paradigms shift, new and similarly delusive paradigms replace them; or as now seems apparent, the new are simply blatant reassertions of the old. Winograd and Flores see this problem most acutely in artificial intelligence and this is a primary reason why they go to Heidegger and Maturana, thinkers who insist on the irreducible complexity of phenomena, for a new approach. If all computer science can do is invent new reductionist gimmicks, then it is locked into a self-enclosed language game where "getting ahead" is defined only parochially or personally, not in terms of larger social effect.

Breakdown represents a jump outside this game, a recognition that it is not product but process, not the gimmick but the getting of the gimmick -- and thus its readiness-to-hand as cognitive "equipment" -- from which value must ultimately derive. To flit across the Internet from idea to idea is convenient and perhaps important in a small way. All things considered, it is nice to have the World Wide Web. More significant, however, are uses of this technology that emphasize the contingency of its structures and claims, that emphasize the experience of "detour," as in Miller's "TRIP," or invite us to think about signs in social context, as in Steadman's "Placing." Narrative as technological breakdown might offer an alternative to a hypnagogic, fetishizing view of language as self-consuming signifier. In this regard it could represent a first step toward the Jamesonian project and thus a possible exit, in the art world at least, from the long meander of postmodernism. It might suggest a way of thinking about information technology that is not tied to mass markets, economies of scarcity, and general solutions. At very least cybertext provides a smart answer to the Control Voice of push media and WebTV, and this might be a toehold for more meaningful resistance. What if the audience pushes back?

However, before we push back this flight to Mr. Barlow's brave new electroworld, some stipulations are in order. To begin with, it is impossible to theorize an end of theory or to define in abstract terms a set of anti-reductionist practices. To describe cybertext as a general method for the critique of general methods is absurd; or as Pynchon says of all metaphor, it is "a thrust at truth and a lie" (129). Any general statement made here will find its limits in the particular. The things we have called cybertexts are too multifarious to be grouped so easily under a single heading. There are major differences in design, approach, and community of reception between the Millers' Myst, Joyce's afternoon, Pavic's Dictionary of the Khazars, and synchronous interactive spaces such as LinguaMOO, though all these fit within Aarseth's general definition of cybertext and all arguably represent designs for breakdown.

To complicate matters further, cybertext is very much work in progress, keyed to the pace and rhythms of the information industries. Especially since the takeoff of the World Wide Web, change in the techne of interactive writing has been relentless. At dizzyingly short intervals, engineers at Netscape Communications release radical new design tools whereupon their counterparts at Microsoft raise the ante, redefining the nature of the game. It might thus seem pointless to consider cybertext as a coherent development or a unified aesthetic field. What is the point of trying to think seriously about something that changes faster than the speed of print? Or more critically, why resist the centripetal movement of media change that is currently sucking the Web toward the singularity of broadcasting?

This question adds a very important complication to the argument. Anxiety about technological displacement is hardly limited to champions of the book. Birkerts' polemic against information culture might reasonably be described as reactive, if not reactionary; but the same could also be said about a defense of hypertext against insurgent neo-television. No party wants its textual totem, be it book or hypertext, to land among postcards and telegrams on the cultural scrapheap. Birkerts takes up his crusade against the electronic word because he fears for the future:

As a writer I naturally feel uneasy. These large-scale changes bode ill for authorship, at least of the kind I would pursue. There are, we know this, fewer and fewer readers for serious works. Publishers are increasingly reluctant to underwrite the publication of a book that will sell only a few thousand copies. But very few works of any artistic importance sell more than that. And those few thousand readers -- a great many of them, it turns out, are middle-aged or older. The younger generation have not caught the habit. (28)

These remarks lend interesting perspective to Birkerts' conception of the word as "broker" of a "transaction" and Fauth's approach to narrative as commodity "content." Indeed, "serious" words invested with aesthetic "quality" are the coin of the publishing realm, the stuff on which one builds a literary (which these days means an academic) career. There may be a legitimate dispute about aesthetics here but there is clearly also a turf war, or as we learned to say in the Reagan years, a lively exploitation of the free market in ideas. On either side of the technology barricades, writers and critics who take their work seriously feel imperiled, as if they are caught up in a dualistic, zero-sum game where one can only gain at the other's expense. There are claims to be defended and converts to win. Readers must be introduced to the "habit" of page turning or link-following while still impressionable -- as if the goods in question were not words but cigarettes. As we confessed at the outset, this is not so much Blakean contrariety as the paralytic logic of the Cold War; and in that respect this essay is as badly tainted as anything on the other side.

Is there an alternative to this mutual hostility? Bolter's notion of "special literacies" might provide some hope, suggesting that network culture could assist and not displace the culture of the book, for instance by helping serious writers build communities of readers. One of the few successful ventures so far launched on the Internet is Amazon Books, a business that offers among other things help in finding obscure and out-of-print titles. Believers in progressive, pluralist uses of technology see no reason why books cannot co-exist with postmodern information systems including various forms of cybertext. Unfortunately, the most committed of the elegists will not accept such compromises. Birkerts rejects information technologies because he considers them insidious, seductive, and totally incompatible with a literary life -- one lived ideally in a world largely free of machines. As he sees it, any benefit that might flow from electronic media is far outweighed by their ill effects. The hardness of this line makes it difficult to share Bolter's optimism about the cultural outlook. There is also the lesson of the 1980s, which saw both the one-sided results of the Cold War and the rapid reorganization of western economies around a shrinking number of major industrial interests. Pluralism, it would seem, is largely a creature of theory.

Instead of harmonious mosaics it is probably more realistic to think about lacunae and margins. Birkerts clearly knows this: after worrying about the diminishing population of readers, he cannily observes that he has enough to see him through the rest of his "time below" (29). Technophobia and cultural conservatism are strong niche markets. Indeed, after a certain amount of shaking-out and downsizing, one can imagine certain areas of the book trade thriving in the next few decades, especially as erosion of the middle class progresses and higher education belongs more exclusively to children of exceptional privilege. As they say in the marketing world, these are good demographics. There may be fewer readers in the future, but their pockets will be deep.

It is harder to read cybertext's Tarot. Its marginal position in the coming empire of signs seems much more tenuous, since it cannot be advertised as the precious relic of a lost regime. The episteme of print may be at a disadvantage when set against film, television, and other technologies of the spectacle; but print (as the elegists define it anyway) shares the ideology of sovereign authorship and stable products, and this common interest is not trivial. By contrast technologies like hypertext are difficult, discontinuous, and inherently unsettling, less committed to sovereignty than discursive unfolding. Compared to the tradition of print on the one hand and the regime of broadcasting on the other, cybertext is something new in the earth. As such it will always be followed by the fatal sentence: "Nothing new lasts."

Dr. Johnson's cynical jibe holds true, however, only if we insist on validating product over process, maintaining an ideology of "content" in which meaning is commodity and, to pervert McLuhan, the medium is the package. This is not the only way to think about language and technologies. To think otherwise, we may need to stop mourning the late age of print and recognize that imaginative work, and even literature, may continue in other contexts. As Pfeil said two decades ago, "writing should demand we see." Part of what we see must be the medium itself, and in the case of electronic text, its dubious, contingent claims to represent a semantic space, broken (happily) out of the seamlessness of cyberspace. Beyond this, Pfeil would no doubt insist that we see something larger, namely the articulation between discursive practices like cybertext and late-capitalist enterprises like software development, publishing, and broadcasting -- that we recognize our alleged obsolescence and evident marginality, or as Pynchon calls it, our preterition. In this regard shifting the aesthetic ground from validation of "quality" to exploration of difference, from the (mere) novelty of product to the production of novelty, could begin a larger process. This enterprise might mean creating markets defined less by demographics (in the root sense of scripted populations) than by populations of scriptors, readers and writers of "special" literacies. It might also mean re-examining certain assumptions about property, scarcity, and the distribution of profit, at least in the domain of information. That would indeed mean change; but as the man says, change is what writing demands.

Works Cited

Aarseth, Espen. "Cybertext: Perspectives on Ergodic Literature." Diss. University of Bergen (Norway), 1995. [Forthcoming from Johns Hopkins UP.]

Barlow, John Perry. "A Declaration of the Independence of Cyberspace." Self-published manifesto, 8 February 1996. 16 pars. Online. WWW. 8 April, 1997.
Go there

Birkerts, Sven. The Gutenberg Elegies: The Fate of Reading in an Electronic Age. Boston: Faber and Faber, 1994.

Bolter, Writing Space: The Computer, Hypertext, and the History of Writing. Hillsdale NJ: Lawrence Erlbaum, 1991.

Borges, Jorge Luis. Labyrinths: Selected Stories and Other Writings. New York: New Directions, 1964.

Coover, Robert. "The End of Books." New York Times Book Review 21 June 1992: 1, 11, 24-25.

De Landa, Manuel. War in the Age of Intelligent Machines. New York: Verso, 1993.

Dillon, A., C. McKnight, and J. Richardson. "Space -- The Final Chapter or Why Physical Representations are not Semantic Intentions." Hypertext: A Psychological Perspective. Ed. C. McKnight, A. Dillon, and C. Richardson. New York: Ellis Horwood, 1993. 169-91.

Douglas, J.Y. "Where the Senses Become a Stage and Reading is Direction: Performing the Texts of Virtual Reality and Interactive Fiction." The Drama Review 37(4) [1993]. 18-35.

Douglas, J.Y. "'How Do I Stop this Thing?' Closure and Indeterminacy in Interactive Narratives." Hyper/Text/Theory. Ed. G.P. Landow. Baltimore: Johns Hopkins UP, 1994. 159-188.

Fauth, Jurgen. "Poles in Your Face: The Promises and Pitfalls of Hyperfiction." Mississippi Review 6 September, 1995. 38 pars. Online. WWW. 8 April 1997.
Go there

Glazier, Loss. "Jumping to Occlusions..." [forthcoming in Postmodern Culture]

Guyer, Carolyn. "Written on the Web." Feed September, 1995. Online. WWW. 14 April 1997.
Go there

Harpold, Terence. "Threnody: Psychoanalytic Digressions on the Subject of Hypertexts." Hypermedia and Literary Studies. Ed. G.P. Landow. Cambridge: MIT Press, 1990. 171-84.

Harpold, Terence. "Conclusions." Hyper/Text/Theory. Ed. G.P. Landow. Baltimore: Johns Hopkins UP, 1994. 189-223.

Jameson, Frederic. Postmodernism Or, the Cultural Logic of Late Capitalism. Durham: Duke UP, 1990.

Johnson, George. "Old View of Internet: Nerds. New View: Nuts." New York Times 30 March, 1997, National ed.: E1+.

Johnson-Eilola, Johndan. Nostalgic Angels: Reconfiguring Hypertext Writing. Norwood NJ: Ablex, 1997.

Joyce, Michael. Afternoon, a Story. Cambridge: Eastgate Systems, 1990.

Joyce, Michael. Of Two Minds: Hypertext Pedagogy and Poetics. Ann Arbor: U. Michigan Press, 1994.

Kelly, Kevin and Gary Wolf. "We Interrupt This Magazine for a Special Bulletin -- PUSH!" Wired 5.03 (March, 1997). 12-23.

Kirschenbaum, M. G. "The Cult of Print." Postmodern Culture 6.1 (1995): 12 pars. Online. WWW. 8 April 1997.
Go there

Kolb, David. Socrates in the Labyrinth. Cambridge: Eastgate Systems, 1995.

Landow, George P. "The Rhetoric of Hypertext." Hypertext '87 Proceedings. Chapel Hill: Association for Computing Machinery, 1987. 331-43.

Landow, George P. Hypertext: The Convergence of Contemporary Critical Theory and Technology. Baltimore: Johns Hopkins UP, 1992.

Lanham, Richard. The Electronic Word: Democracy, Technology, and the Arts. Chicago: U. Chicago Press, 1993.

Liu, Alan. The Voice of the Shuttle. Web site inaugurated 1994. Online. WWW. 14 April 1997.
Go there

McGann, Jerome. "The Rationale of Hypertext." Essay published 6 May 1995. Online. WWW. 14 April 1997.
Go there

Miller, Matthew. "TRIP." Hypertext fiction. Postmodern Culture 7.1 (1996). Online. WWW. 11 April 1997.
Go there

Moulthrop, Stuart. "Traveling in the Breakdown Lane: A Principle of Resistance for Hypertext." Mosaic 28(4) [1995]: 55-77.

Moulthrop, Stuart and Nancy Kaplan. "Where No Mind Has Gone Before: Ontological Design for Virtual Spaces." Proceedings of the ACM Hypertext Conference. Edinburgh: Association for Computing Machinery, 1994. 212-23.

Nelson, Theodor H. Literary Machines. Sausalito: Mindful Press, 1990.

Nelson, Theodor H. "Where the Trail Leads." A Symposium in Honor of Vannevar Bush on the 50th Anniversary of "As We May Think." Massachusetts Institute of Technology. Cambridge. 12 October 1995.

Page, Barbara. "Women Writers and the Restive Text: Feminism, Experimental Writing and Hypertext." Postmodern Culture 6(2) [1996]. 29 pars. Online. WWW. 14 April 1997.
Go there

Pfeil, Fred. Another Tale to Tell: Politics and Narrative in Postmodern Culture. New York: Verso, 1990.

Pynchon, Thomas. The Crying of Lot 49. New York: Lippincott, 1966.

Rheingold, Howard. The Virtual Community: Homesteading on the Electronic Frontier. Cambridge: Addison Wesley, 1994.

Röscheisen, M., C. Mogensen, and T. Winograd. "Beyond Browsing: Shared Comments, SOAPs, Trails, and On-Line Communities." 1995 World Wide Web Conference. Online. WWW. 12 April 1997.
Go there

Rosenberg, Jim. "Navigating Nowhere/Hypertext Infrawhere." SIGLINK Newsletter 3 (December, 1994). Online. WWW. 13 April 1997.

Rosenberg, Jim. "The Structure of Hypertext Activity." Hypertext '96 Proceedings. Washington: Association for Computing Machinery, 1996. 22-30.

Shumate, Michael. "Hyperizons." Annotated index of hypertext fictions. Online. WWW. 16 April 1997.
Go there

Steadman, Carl. "Placing." Hypertext fiction. On-line. WWW. 12 April 1997.
Go there

Tolva, John. "Ut Pictura Hyperpoesis: Spatial Form, Visuality, and the Digital Word." Hypertext '96 Proceedings. Washington: Association for Computing Machinery, 1996. 66-73.

Ulmer, Gregory. Heuretics: The Logic of Invention. Baltimore: Johns Hopkins UP, 1994.

Winograd, Terry and Fernando Flores. Understanding Computers and Cognition: A New Foundation for Design. Cambridge: Addison Wesley, 1986.


1. Birkerts, 78. [Back]

2. Pfeil, 28. [Back]

3. Though Fauth complains specifically about "hyperstories" on the Internet, he offers no detailed consideration of any such. Only Michael Joyce's afternoon receives extensive discussion. Even assuming Fauth's opinions about this work are well founded (though they in fact say very little about its "content"), they are arguably irrelevant. Afternoon was written before the inception of the World Wide Web, is not available on-line, and differs in notable respects from projects like "Girl, Birth, Water, Death," "TRIP," "Placing," and numerous other Web fictions. Fauth does mention a few Internet texts, including "Stories from Downtown Anywhere" and "Hypertext Hotel," but rules them out because they are collaborations, where presumably "quality" is less important. For a much more careful (though hardly less critical) survey of Web fiction, see Guyer. [Back]

4. As Birkerts notes in the introduction to Gutenberg Elegies (p.4), attitudes toward current information technologies do not map onto the old liberal/conservative axis. Newt Gingrich and Timothy Leary have both been advocates of the Internet. By the same token, for all his dislike of late capitalism, Birkerts stands in the same ideological corner as William Bennett and E.D. Hirsch. I am interested less in old ideological positions than in those now emerging which may be defined more by atttitudes toward information and interpretive authority than by traditional political concerns. [Back]

5. Outside the computer game industry there were until recently two major publishers of electronic writing in the United States: Voyager Company and Eastgate Systems. Voyager recently announced its departure from the business. [Back]

6. Following the foundational studies by Bolter (1991) and Landow (1992), important commentary has come from Coover (1992), Lanham (1993), Joyce (1994), Ulmer (1994), Liu (1994), and McGann (1995). Essays by Douglas (1993, 1994), Harpold (1990, 1994), Rosenberg (1995, 1996), Page (1996) and Tolva (1996) have contributed insights into cybertextual phenomenology and reader response. Recent book-length studies from Aarseth (1995) and Johnson-Eilola (1997) help place electronic forms in social and historical context. Kolb's hypertext (1995) represents the first extensive use of the medium for critical discourse. This list is hardly exhaustive. [Back]

7. All Web browsers differentiate visited from unvisited destinations by changing the highlighting color of cue material. Most also display the destination code for links (usually a Uniform Resource Locator) in small type somewhere on the screen. This information is at best a cryptic guide to what lies on the other end of the link. [Back]